[dazq]

Secondary serial protocol is one of request /reply, so the speeduino would request the value over serial3 and the O2 control reply back. The request frequency is adjustable.

The speeduino can interface is simple setup to periodically ( adjustable again) listen out for the canid that carries the O2 data and reads it.
The interface sends the data to the speeduino (or if it's teensy can then we'll it's already there! )

So the O2 device only need respond on demand from serial or broadcast on canbus.

BUT
What we don't yet have in speeduino firmware is a way to select the external O2 data value to use by speedy, you could display and log it but can't yet use it.

Ps if you need any guidance with our serial protocol or can protocol then let me know

[Alfagta]

I'm curious to this, as I would think in SERIAL (not CAN!), that the controller would continuously puke it's data out to Speeduino. Speeduino would then read the latest received data for calculating when it wants to. No requests, no waiting, instant read of latest data with no delay. Yes? No?
[EDIT - this is similar function to analog reading, where the data is constantly available for Speeduino to instantly read when ready for it.]

David

Serial broadcast would be possible, for sure.
But....
what if you have closed loop lambda. And after a Lambda 0.95 reading (target 1.0) the lambda controller stops sending data to the Speeduino.
Speeduino needs to error check this, otherwise speeduino leans out the fuel until the engine fails.?
Best Regards

[dazq]

There is a flag set if the serial device fails to respond to speeduino data request, ie timeout.
Again this value is not currently checked but could be as part of the implementation of using external data sources

[plazomat]

[/quote]
Serial broadcast would be possible, for sure.
But....
what if you have closed loop lambda. And after a Lambda 0.95 reading (target 1.0) the lambda controller stops sending data to the Speeduino.
Speeduino needs to error check this, otherwise speeduino leans out the fuel until the engine fails.?
Best Regards
[/quote]

Isn't there a setting somewhere in TS that limits the amount of correction the closed loop can make? or is that a MS thing?

Plaz

[PSIG]

Isn't there a setting somewhere in TS that limits the amount of correction the closed loop can make? or is that a MS thing?

Correct, it actually has two limit-safeties. The obvious one is right in the O2 settings for control authority limit, and for example I have mine generally set at 5% to 7% maximum, as if needs more than that then something is wrong with the sensor, controller, fuel, or something else that's out to lunch. I also set the gauge to light full-red that it's getting to the limit of what it should need, and that I need to back-off and find what's going wrong.

For a second safety, I limit the Lambda/AFR range it is allowed to correct within, to the limits of what that engine should ever see. A general range might be 11:1 to 17:1 (Lambda .75 to 1.15), preventing correction if (again) something is not right. Something not right could be as simple as the O2 sensor not yet up to temperature, and I don't want it pulling fuel completely out of control.

David

[Groopan]

Hey guys I made my own version of this shield using blyund's schematics. There were several corrections I had to make most of them my fault and I noticed that the driver chip needed to be grounded or else I get intermittent heater diagonistics errors. So I got the sensor to calibrate successfully and then the heating starts it appears to heat up successfully blinking and ramping up voltage from 2 to 12v in the specified time then it switches over to pwm and the heater light goes steady and the measured voltage seems to be around 6-7v in air .
What I have noticed is during heat up the sensor will get hot till it starts to glow a little then shuts off when the heater control is passed to pid .

The readings I get after this seems weird I get a oxygen % of 6-7 in air .the heater optimal value is 174 counts and it seems to be maintaning this setpoint well enough.
The lambda value is around 1.5.

Any one can guide me as what can I check again . I will attach pics of my shield and results so far.

Ok so I measured the current that goes to iP on the cj125 and it's 0.7xx amps which matches the 6% so I am unsure what is happening may be the sensor is not hot enough but the cj125 says the temp is 147 counts which is the opitumum temp value . Humm any ideas?

[ruzki]

@ groopan
soory to say but your layout seems to be very wrong. It looks like do not have a proper ground plane etc.
Funny things can happen if you have ground noise.
You could post your schematic and layout.
Do you have a mosfet driver to drive the mosfet??

[Groopan]

@ groopan
soory to say but your layout seems to be very wrong. It looks like do not have a proper ground plane etc.
Funny things can happen if you have ground noise.
You could post your schematic and layout.
Do you have a mosfet driver to drive the mosfet??

I have used blyund's desigh I admit this is my first time designing a pcb from scratch I used jclpcb there were a few mistakes the mosfet driver was not grounded . And some vbat were not connected I have corrected these and have reached a point where the sensor calibrates well but once I start to read o2 things go south . Temp control seems to work well though.
Wouldn't a grounding problem cause irattic readings.

Any advice will be grately appreciated.

https://easyeda.com/Groopan/shield-uno

[tooly]

layout with atmega32m1 is done. double layer on side components placement 25x60mm.

[Groopan]

@ groopan
soory to say but your layout seems to be very wrong. It looks like do not have a proper ground plane etc.
Funny things can happen if you have ground noise.
You could post your schematic and layout.
Do you have a mosfet driver to drive the mosfet??

So today I manually changed the setpoint for the UR value to 90 and noticed I started to read 20% o2 strange cal optimum setpoint says 174 counts heater does not glow may be it's the mosfet I choose HUFA75321D3 I drive this with a MCP14E9. But it does meet the setpoint and holds it so I am not to sure what's the deal .